cs.CVNov 3, 2026

Mitigating Spatial Blindness in Multiple Instance Learning for Whole-Slide Image Analysis

Authors: Xiangyu Li, Jincheng Li, Xing Xie, Bin Guan, Shaohua Kevin Zhou, Rui Yan, Ran Su

Organizations: College of Intelligence and Computing, Tianjin University, Tianjin, China · Shanghai Artificial Intelligence Laboratory, Shanghai, China · Medical Imaging, Robotics, Analytic Computing & Learning (MIRACLE) Lab, USTC Suzhou Institute for Advanced Research, Suzhou, China · School of Biomedical Engineering, University of Science and Technology of China, Hefei, China · 

Abstract

Whole-slide multiple instance learning can predict diagnostic and prognostic labels from tissue composition while making limited use of spatial organization. Providing coordinates or a graph does not establish that a trained predictor uses tissue neighborhoods. We examine this gap as spatial blindness and hypothesize that early composition learning can weaken supervision of a spatial pathway. We introduce ResTopoMIL, which freezes a learned composition predictor before training a graph residual with a coordinate-reassignment separation loss. Permutation and information analyses distinguish architectural symmetry, available spatial information and the proposed optimization mechanism. On a controlled spatial task, staged training achieved an AUC of 0.987, compared with 0.684 for the same jointly trained architecture. Performance remained high on a composition-only task. Across four WSI classification and five survival benchmarks, ResTopoMIL improved mean primary performance on eight tasks; four improvements remained significant after Holm-adjusted patient-cluster tests. Coordinate-free residual controls recovered some of the gains, while neighborhood destruction reduced performance close to the composition anchor. Neighborhood-preserving transformations and anatomy-stratified perturbations further characterized this dependence. These results support fixing a composition reference as a practical intervention for learning spatial corrections, while retaining additional appearance capacity as a contributing explanation.

Figures & tables

Explore similar work

CardsList
  1. Spatial Blindness in Whole-Slide Multiple Instance Learning

    May 17, 2026Xiangyu Li, Ran SuHistopathology Image ClassificationMultiple Instance Learning

  2. Jigsaw Regularization in Whole-Slide Image Classification

    Mar 20, 2026So Won Jeong, Veronika RočkováHistopathology Image ClassificationWhole-Slide Image Analysis